Markov Networks for Super-Resolution

    •  William T. Freeman, Egon C. Pasztor, "Markov Networks for Super-Resolution", Tech. Rep. TR2000-08, Mitsubishi Electric Research Laboratories, Cambridge, MA, March 2000.
      BibTeX TR2000-08 PDF
      • @techreport{MERL_TR2000-08,
      • author = {William T. Freeman, Egon C. Pasztor},
      • title = {Markov Networks for Super-Resolution},
      • institution = {MERL - Mitsubishi Electric Research Laboratories},
      • address = {Cambridge, MA 02139},
      • number = {TR2000-08},
      • month = mar,
      • year = 2000,
      • url = {}
      • }
  • Research Areas:

    Artificial Intelligence, Computer Vision, Machine Learning


We address the super-resolution problem: how to estimate missing high spatial frequency components of a static image. From a training set of full- and low- resolution images, we build a database of patches of corrsponding high- and low-frequency image information. Given a new low-resolution image to enhance, we select from the training data a set of 10 candidate high-frequency patches for each patch of the low-resolution image. We use compatibility relationships between neighboring candidates in Bayesian belief propagation to select the most probable candidate high-frequency interpretation at each image patch. The resulting estimates of the high-frequency image are good. The algorithm maintains sharp edges, and makes visually plausible guesses in regions of texture.